Researchers do not owe their subjects the same level of care that physicians owe patients, but they owe more than merely what the research protocol stipulates. In keeping with the dynamics of the relationship between researcher and subject, they have limited but substantive fiduciary obligations.

By virtue of their professional ethics as healers and because of their specialized technical knowledge and clinical experience in assessing and reacting to real and potential emergencies, physicians have an obligation to offer an unsolicited medical opinion when the following conditions are met: (1) physicians assess a high probability of potentially serious disease in a stranger because of information presented to them, either in the form of a communication or physical signs; (2) physicians judge this information to be latent (not (...) readily interpretable as potentially dangerous by the stranger) and likely to remain latent prior to the onset of symptoms; (3) the physicians possess the medical knowledge appropriate to the professional interpretation of this information. Although not a morally risk-free endeavor (invasion of privacy and the potential creation of a ‘sick role’ whether or not the diagnosis is correct), offering an unsolicited medical opinion under the above conditions can prevent suffering and save lives in unsuspecting strangers. Keywords: bystander, professional duty, patient-physician relationship, privacy, altruism * I would like to dedicate this essay to my father, Martyn C. Ratzan, M.D., who taught me never to listen to whispers too close to the window. CiteULike Connotea Del.icio.us What's this? (shrink)

: Bioethicists have failed to understand the pervasively paternalistic character of research ethics. Not only is the overall structure of research review and regulation paternalistic in some sense; even the way informed consent is sought may imply paternalism. Paternalism has limits, however. Getting clear on the paternalism of research ethics may mean some kinds of prohibited research should be reassessed.